A protocol for the annotation of evaluative stance and metaphor across four discourse genres

Abstract

The present article contributes to research on evaluation by addressing two complementary objectives: first, we present a protocol for the identification and annotation of evaluation in English discourse and, second, we show the results of the implementation of the protocol in the annotation of evaluation in a sample of a corpus of four genres. We first describe the protocol by discussing the theoretical and methodological grounding of the annotation scheme, the criteria, the categories, the steps for the implementation of the protocol and an illustrative example of the application of the protocol to a short extract. We subsequently provide the preliminary results of a pilot study with the frequency of evaluative expressions across the four genres. Results show that while adjectives and non-metaphoric evaluative expressions are overall more frequent, there are differences regarding the preference for positive or negative value and regarding the frequency of function.